AI Paper Atlas

Research map / Video / Video segmentation and tracking

MEMC-Net: Motion Estimation and Motion Compensation Driven Neural Network for Video Interpolation and Enhancement

Wenbo Bao, Wei‐Sheng Lai, Xiaoyun Zhang, Zhiyong Gao, Ming–Hsuan Yang

TPAMI 2019 · Video › Video segmentation and tracking · approach: Video frame interpolation · 375 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add mask estimation to flow-based model (ablation)MEMC-Net (bilinear kernel, no post-proc) · Vimeo90KPSNR: 32.73 -> 32.94Improved
add post-processing network to flow-based model (ablation)MEMC-Net (bilinear kernel, no mask) · Vimeo90KPSNR: 32.73 -> 33.25Improved
add both mask estimation and post-processing (ablation)MEMC-Net (bilinear kernel) · Vimeo90KPSNR: 32.73 -> 33.40Improved
replace fixed bilinear kernel with learned adaptive kernel (ablation)MEMC-Net (mask and post-proc) · Vimeo90KPSNR: 33.40 -> 34.02Improved
replace bilinear kernel with learned kernel without mask (ablation)MEMC-Net (post-proc, no mask) · Vimeo90KPSNR: 33.25 -> 33.29No clear effect
fine-tune the flow estimation network instead of fixing it (ablation)MEMC-Net · Vimeo90KPSNR: 32.05 -> 34.02Improved
add flow estimation network to kernel-only model (ablation)MEMC-Net (kernel only, post-proc) · Vimeo90KPSNR: 33.31 -> 34.02Improved
add flow estimation network to kernel-only model (ablation)MEMC-Net (kernel only, post-proc) · UCF101PSNR: 34.97 -> 34.95No clear effect
use parallel flow and kernel estimation instead of sequential model (ablation)MEMC-Net · Vimeo90KPSNR: 32.94 -> 34.02Improved
use full model instead of small SPyNet-based model (ablation)MEMC-Net_s to MEMC-Net · Middlebury Other setinterpolation error: 2.44 -> 2.24Mixed
add context information and residual blocksMEMC-Net to MEMC-Net* · Vimeo90KPSNR: 34.02 -> 34.40Improved
use multi-frame MEMC-Net_DN instead of single-frame EDSR_DNvideo denoising · Vimeo90KPSNR: 35.11 -> 36.35Improved

Later papers that compared with this method

5 later accepted papers report MEMC-Net* as a baseline in their results tables; 5 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy